0
\$\begingroup\$

I have a scenario where I would like to use N-channel MOSFETs, because they have lower on-resistance and are cheaper.

schematic

simulate this circuit – Schematic created using CircuitLab

Would there be any problems with this approach?

I've read this answer but I understand the main concern is somehow accidentally forming a closed loop. But if the MOSFET is safely at the negative terminal of each voltage source, I don't see how there would be any risk?

Am I correct that the positive terminal of either voltage source cannot conduct until the negative terminal is also connected? In effect, as if it wasn't connected at all?

\$\endgroup\$
1
\$\begingroup\$

Disconnecting the negative side is sufficient provided that the MOSFET is the only connection to the negative side of the supply. I have designed electronic circuit breakers for batteries that work that way. They worked fine and were even certified for use in an explosive atmosphere.

\$\endgroup\$
4
\$\begingroup\$
  1. My original thought was that your N channel MOSFETs are upside down because of the way you have the on/off switch arrangement. However, after @ThePhoton pointed out my blindness, another problem arose in the "turning off" area. To turn-off the MOSFET you do need to have a different arrangement for the "switch" you have drawn. The disconnected pin on SW3 (or SW4) has to be connected to the source terminal on the MOSFET or it won't properly shut down "off" - they'll conduct with 1 or 2 volts across them.
  2. Generally most MOSFETs are OK with a 12 volt gate-source signal but some may not be.
  3. Presumably your batteries are safe to be used in parallel.

Other than that, I'd have no problem doing it this way providing there isn't an electrical path from the negative battery terminal going somewhere else that you have not shown in your picture. The extra connection due to the wrongly connected switch will of course be another connection so some care is needed here to make sure it works correctly and doesn't leak current and discharge the battery.

Is it sufficient to disconnect the negative terminal of a voltage source?

It's sufficient in many applications.

\$\endgroup\$
  • \$\begingroup\$ I think this is a scenario where you want the FET "upside-down". Because current will actually be flowing from the "bottom" to the "top", and you don't want the body diode to conduct. (Or, put another way, it might be more clear to draw the FET below the chassis ground rail) \$\endgroup\$ – The Photon Jan 3 '17 at 17:31
  • \$\begingroup\$ @thephoton yes you are quite right I shall make amends for my shortsightedness. However to turn the fet off requires the switch to be repositioned with the off contact connected to source. \$\endgroup\$ – Andy aka Jan 3 '17 at 17:54
  • \$\begingroup\$ Yeah I had a feeling there was an issue with trying to use a "disconnected" source to drive the gate (I was thinking about how to turn it on)...but wasn't seeing the best solution. \$\endgroup\$ – The Photon Jan 3 '17 at 18:19
1
\$\begingroup\$

Am I correct that the positive terminal of either voltage source cannot conduct until the negative terminal is also connected? In effect, as if it wasn't connected at all?

Yes, but you have to watch out for all possible routes. A common problem with trying to power down part of a system is "phantom power": a "signal" line which is held high ends up powering the device. Similarly if you disconnect the power supply ground but forget about grounded signals, cable shields tied to ground, etc., then you will have problems.

It doesn't look like this applies in your scenario because the MOSFET is right next to the power supply and will disconnect all possible routes to that negative terminal.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.